Reaction Time Test

How to use

1. Click or tap the box, or press Space. It turns red.
2. Wait. After a random 1.5 to 4.5 seconds it turns lime. Click as fast as you can.
3. Do five rounds. You get the average, your best round, and a comparison with a published reference.

What is a good reaction time?

For context, Human Benchmark reports a median of 273 ms for its visual reaction test. That is a reference from another website, not a measured distribution of ReactionCheck users. Your device, input method and test conditions affect the result, so a lower score does not by itself establish athletic ability or gaming skill.

Your score includes display, audio and input delays as well as your response. A 60 Hz frame interval is about 16.7 ms; at 144 Hz it is about 6.9 ms. These intervals are not the complete device latency: panel processing, input hardware and browser scheduling add varying delays. Compare repeated sessions on the same device and input method rather than subtracting a fixed hardware allowance.

Why the F1 test is different

In our F1 reaction test, five lights come on one per second and go out after a random 0.2–3 second hold. React to lights going out instead of a panel turning green. This is a browser simulation: clicking a mouse cannot reproduce a driver releasing a clutch or a car starting to move. Compare repeated sessions on the same setup.

How to improve reaction time

Use the same device and input method for each session, take breaks, and keep distractions low. A few practice rounds help you learn the controls. Avoid anticipating the signal: a lower number from guessing does not measure a faster reaction. This tool is for practice and comparison, not medical assessment.

Keep display and input settings consistent. Power-saving modes, background work and a change of mouse, keyboard or audio output can change your result. An equipment change does not by itself show that your reactions improved.

How this test works

Each round draws a fresh random delay between 1500 and 4500 ms. When the panel switches to lime the page records performance.now(), a monotonic clock whose precision may be reduced by browser privacy settings, and records it again on your input. The difference is your time. Input is captured on pointerdown, so the clock stops the moment your finger goes down, not when the click completes on release. Space and Enter are captured on key down for the same reason.

Clicking early cancels the round. Your score is the mean of five completed rounds; “best” is the fastest single round. For the visual test we show the difference from Human Benchmark’s published 273 ms median. We do not collect population data or calculate a validated percentile. The browser clock starts before the physical display or speaker finishes presenting the signal, so equipment latency is included.

All tools on ReactionCheck

• Controller testerBrowser-reported buttons, sticks and triggers, with drift and rumble checks where supported.
• FPS testBrowser frame rate, 10 s average, peak rate and a motion clarity strip.
• Spacebar counterPresses per second in 5, 10, 30 or 60 s, with your best saved.
• Mouse testEvery button, the scroll wheel, accidental double clicks and polling rate.
• Dead pixel testFullscreen colours to find dead or stuck pixels, plus a stuck pixel fixer.
• F1 reaction testFive start lights, random hold, react when they go out.
• Audio reaction testReact to a 1 kHz beep instead of a colour change.

Why these tools: they are the checks a gamer or anyone fixing a PC or Mac ends up doing one after another, with tabs full of ad-heavy sites. Everything here runs in the page, measures what it claims with browser APIs, and says plainly where the browser limits the measurement.